Claudin expression is dysregulated in prostate adenocarcinomas but does not correlate with main clinicopathological parameters

Pathology (February 2011) 43(2), pp. 143–148

ANATOMICAL PATHOLOGY

Claudin expression is dysregulated in prostate adenocarcinomas but does not correlate with main clinicopathological parameters CLA´UDIA MALHEIROS COUTINHO-CAMILLO*, SILVIA VANESSA LOURENC¸Oz, FRANCISCO PAULO FONSECA{ AND FERNANDO AUGUSTO SOARES*z

DA

*Department of Anatomic Pathology and {Division of Urology, Department of Pelvic Surgery, Hospital A. C. Camargo, Sa˜o Paulo, and zDiscipline of General Pathology, Dental School, University of Sa˜o Paulo, Sa˜o Paulo, Brazil

Summary Aims: Claudins, a large family of essential tight junction (TJ) proteins, are abnormally regulated in human carcinomas. The aim of this study was to determine the expression of claudins 1, 2, 3, 4, 5, 7, and 11 in prostate samples from Brazilian patients and correlate it with the clinicopathological features of prostate cancer. Methods: Using a tissue microarray (TMA) of specimens of prostate adenocarcinoma and benign prostatic hyperplasia (BPH) we analysed the expression of claudins 1, 2, 3, 4, 5, 7, and 11 by immunohistochemistry. Results: Claudin 4 was down-regulated and claudins 2, 3, and 5 were overexpressed in prostate adenocarcinomas compared with BPH samples. Expression of claudins 1 and 7 was similar in tumours and BPH samples. Claudin 11 was absent from all prostate samples. Overexpression of claudin 3 was associated with perineural invasion ( p ¼ 0.014) and tended to occur in advanced stages of the disease ( p ¼ 0.064). Increased expression of claudin 5 was marginally associated with perineural invasion ( p ¼ 0.060). Conclusions: Our results suggest that alterations in claudin expression occur in prostate cancer cells, although we have not found an association with the main clinicopathological parameters. Key words: Claudins, immunohistochemistry, prostate cancer, tissue microarray. Received 9 January, revised 4 November, accepted 9 November 2010

INTRODUCTION Prostate cancer is the most commonly diagnosed malignancy and the second-leading cause of death from cancer in men in the United States and many Western countries.1 It is also highly prevalent in Brazil, where it had an estimated occurrence of 49 530 new cases of prostate cancer in 2008.2 There are no definitive markers that distinguish aggressive from indolent prostate cancers or that predict outcomes after definitive therapies. Several events effect cancer development and studies have proposed that adhesion molecules regulate disease progression.3 The tight junction (TJ) is part of the apical junctional complex and is intimately involved in paracellular permeability and cell polarity.4 These junctions comprise several membrane and peripheral proteins; primarily occludins, zonula occludens (ZO), and claudins.5 Due to their ability to recruit signalling proteins, tight junction proteins have also Print ISSN 0031-3025/Online ISSN 1465-3931 DOI: 10.1097/PAT.0b013e3283428099

#

been hypothesized to control proliferation, differentiation, and other cellular functions.3,4 The claudin family of proteins regulates the formation and function of tight junctions.6 There are at least 23 human claudin family members, which are expressed in epithelial and endothelial cells. Claudins are integral transmembrane proteins that have four membrane-spanning regions and range in molecular weight from 20 to 27 kDa.7 The sealing of tight junctions by claudins is likely to be mediated in part by phosphorylation. The cytoplasmic C-terminus of claudins contains not only several phosphorylation sites but also a PDZ-binding motif, to which PDZ domain-containing proteins, such as the tight junction proteins ZO-1, -2, and -3, bind.7 Altered claudin expression has been observed in various human epithelial cancers, including colorectal,8,9 pancreatic,10,11 breast,12,13 and ovarian cancer.14,15 Recently, studies addressed the role of claudins in prostate adenocarcinoma, but variability in their results has left this issue unresolved.16-19 The aim of this study was to supplement this area of research by evaluating the expression of claudins 1, 2, 3, 4, 5, 7, and 11 in prostatic adenocarcinoma and hyperplasia in Brazilian patients and correlate these findings with the clinicopathological hallmarks of prostate cancer.

MATERIALS AND METHODS Tissue samples Paraffin-embedded tissue samples from 227 patients were obtained from the Department of Pathology, Hospital A. C. Camargo, Sa˜o Paulo, Brazil. Their clinical and histological properties are listed in Table 1. Pre-operative prostate specific antigen (PSA) level was available for all patients, and PSA non-failure was defined as PSA levels that remained below 0.2 ng/mL after radical prostatectomy. Biochemical recurrence was defined as the first instance of an increase in PSA above 0.2 ng/mL. The Institutional Ethics Committee approved this study. The cases comprised acinar prostate adenocarcinoma (n ¼ 210) and benign prostatic hyperplasia (BPH) tissue (n ¼ 66), all of which came from prostatectomy specimens. Tissue microarray (TMA) For TMA construction, haematoxylin and eosins (H&E) sections of each case were analysed, and a representative area of the tumour or BPH was selected. The tissues that corresponded to these areas were sampled from the donor block using a tissue microarrayer (Beecher Instruments, USA). Each sample was arrayed once with a 1.0 mm diameter core, spaced 0.2 mm apart. After the arraying was completed, TMA blocks were sectioned at thicknesses of 4 mm. Immunohistochemistry The expression of claudins 1, 2, 3, 4, 5, 7, and 11 was evaluated in prostate tissue samples on a tissue microarray (TMA). The immunostains were performed on

2011 Royal College of Pathologists of Australasia

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

144

Pathology (2011), 43(2), February

COUTINHO-CAMILLO et al.

Table 1 Summary of the clinicopathological characteristics of the prostate cancer patients Characteristic

Category

No. patients

Gleason score

4–6 7 8–9 I/II III/IV No Yes No Yes No Yes No Yes No Yes

116 66 26 97 107 36 171 174 30 156 35 185 7 141 64

Clinical stage Perineural invasion Seminal vesicle invasion Extraprostatic extension Lymph node metastasis Biochemical recurrence

duplicate tissue slides; duplicate sections were 40 mm apart. The slides were deparaffinised, rehydrated, and then subjected to antigen retrieval. Tissue sections were incubated at room temperature for 2 h with the primary antibody. Details on antigen retrieval methods as well as primary antibodies clones, source, and working solutions are given in Table 2. The antigen-antibody complexes were visualised using the Advance detection system (Dako, USA). Positive controls were used according to the manufacturer’s recommendation. Semi-quantitative analysis of the results was performed using a conventional optical microscope. The semi-quantitative analysis considered the following scores based on the evaluation of area stained in each core (staining intensity was not considered for the final score): focal expression, when up to 30% of the total area of the prostatic tissue (neoplastic or BPH) was stained by the marker; and overexpression (widespread), for staining of more than 30% of the prostatic tissue, adapted from Sheehan et al.16 Statistical analysis Analysis of the association between claudin levels and the demographic and clinicopathological characteristics of the patients was performed using the chisquare test. The following parameters were included: Gleason score (Gleason score 6, Gleason score ¼ 7, Gleason score 8), perineural invasion (yes or no), seminal vesicle invasion (yes or no), extraprostatic extension (yes or no), lymph node metastasis (yes or no), stage (I/II versus III/IV), and biochemical recurrence (yes or no). Disease-free survival and overall survival probabilities were calculated using the Kaplan-Meier method. The significance level was 5% for all tests. Statistical analyses were performed using SPSS software 13.0 (SPSS, USA).

Table 3

Table 2

Primary serum, clones, source, working titre and antigen retrieval

Primary serum Claudin Claudin Claudin Claudin Claudin Claudin Claudin

1 2 3 4 5 7 11

Clone

Source

Working title

Polyclonal Polyclonal Polyclonal Polyclonal Polyclonal Polyclonal Polyclonal

Zymed, USA Neomarkers, USA Neomarkers, USA Neomarkers, USA Neomarkers, USA Zymed, USA Neomarkers, USA

1:400 1:400 1:500 1:200 1:1000 1:800 1:800

Antigen retrieval Citrate, Citrate, Citrate, Citrate, Citrate, Citrate, Citrate,

pH pH pH pH pH pH pH

6.0 6.0 6.0 6.0 6.0 6.0 6.0

RESULTS Nearly all of the claudins that we tested were expressed in the prostate samples included in the study and in all positive controls; stain intensity was similar in all positive cases. Claudin 11 was not detected in any sample. The results are described in Table 3. Claudins 1, 2, 3, 4, 5, and 7 showed variable expression in the prostate adenocarcinoma and BPH samples. All claudins were expressed on the membranes of epithelial neoplastic cells and controls (Fig. 1). Claudin 1 and claudin 7 were expressed equally between neoplastic and BPH samples. Increased expression of claudins 2, 3, and 5 was observed in the neoplastic samples compared with BPH samples ( p ¼ 0.019, p < 0.001 and p < 0.001, respectively). Claudin 4 expression was decreased in the neoplastic samples ( p ¼ 0.007) compared with BPH samples (Table 3). Expression of claudins was compared with clinicopathological parameters in the neoplastic samples, wherein we found that increased expression of claudin 3 was associated with the presence of perineural invasion ( p ¼ 0.014). Elevated claudin 5 expression was marginally associated with perineural invasion ( p ¼ 0.060). Increased claudin 3 levels were marginally associated with advanced stages of disease ( p ¼ 0.064). We could not analyse the relationship between claudin expression and lymph node metastasis due to the few cases that presented with it (7 cases with lymph node metastasis versus 185 without). Claudin expression in neoplastic prostate tissue had no statistically significant relationship with other parameters (Gleason grade, seminal vesicle invasion, extraprostatic extension, and biochemical recurrence) (Table 4).

Semiquantitative expression of claudins in neoplastic and BPH lesions Expression of claudins, n (%)*

Adenocarcinoma Focal expression Over-expression Total BPH samples Focal expression Over-expression Total Expression status p value

Claudin 1

Claudin 2

Claudin 3

Claudin 4

Claudin 5

Claudin 7

199 (95.7) 9 (4.3) 208

78 (42.9) 104 (57.1) 182

57 (27.7) 149 (72.3) 206

48 (23.2) 159 (76.8) 207

60 (29.1) 146 (70.9) 206

193 (93.2) 14 (6.8) 207

62 (98.4) 1 (1.6) 63 Equally expressed 0.312

34 (60.7) 22 (39.3) 56 Increased in neoplastic 0.019

48 (73.8) 17 (26.2) 65 Increased in neoplastic <0.001

5 (7.8) 59 (92.2) 64 Decreased in neoplastic 0.007

35 (53.0) 31 (47.0) 66 Increased in neoplastic <0.001

63 (98.4) 1 (1.6) 64 Equally expressed 0.112

*

Claudin 11 was not expressed in the prostate samples analysed. Bold p values statistically significant.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CLAUDIN EXPRESSION IN PROSTATE ADENOCARCINOMAS

145

Fig. 1 Expression of claudins in prostate adenocarcinoma samples. (A) Presence of claudin 1 surrounding cytoplasmic membrane of neoplastic cells in contact with tumour stroma. (B) Prominent expression of claudin 2 in all cells that constitute the neoplastic ducts. (C,D) Expression of claudins 3 and 4, respectively, presenting similar patterns: a marked expression surrounding the entire cytoplasmic membrane of neoplastic cells that constitute the ductal structures. (E) Widespread expression of claudin 5 in all neoplastic cells. (F) Delicate expression of claudin 7, especially located at the basal pole of neoplastic cells and at intercellular contacts. (G) Absence of claudin 11 in neoplastic ductal structures.

Overall survival did not differ significantly between the cases that had focal expression or overexpression of any of the claudins.

DISCUSSION Claudins, a large family of essential TJ proteins, are abnormally regulated in human carcinomas, suggesting that they can be used to evaluate neoplastic lesions. Using a TMA that comprised elements of neoplastic and BPH tissue, our study

demonstrated different claudin expression profiles in prostate carcinoma. Tumorigenesis is accompanied by changes in adhesion, such as TJ disruption and the subsequent loss of cell-cell adhesion. These events can mediate the loss of cell differentiation and uncontrolled cell proliferation, leading to the cell-invasive phenotype in malignant neoplasms. Claudins have been proposed to regulate this complex process. We observed a marked loss of claudin 4 expression in prostate carcinoma compared with BPH tissue. This result

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

107

171

36

64

141

35

156

30

Focal expression Over-expression Focal expression Over-expression

Focal expression Over-expression Focal expression Over-expression

Focal expression Over-expression Focal expression Over-expression

Focal expression Over-expression Focal expression Over-expression

Focal expression Over-expression Focal expression Over-expression

Focal expression Over-expression Focal expression Over-expression Focal expression Over-expression

Expression level*

34 2 164 p ¼ 0.657 7

135 6 61 p ¼ 0.889 3

149 7 34 p ¼ 0.664 1

167 7 28 p ¼ 0.515 2

94 3 101 p ¼ 0.382 6

111 5 63 p ¼ 0.989 3 25 1

Claudin 1 expression

16 13 62 p ¼ 0.144 91

56 65 21 p ¼ 0.203 37

57 76 15 p ¼ 0.788 18

64 90 13 p ¼ 0.421 13

33 47 43 p ¼ 0.724 55

45 54 21 p p ¼ 0.195 40 12 10

Claudin 2 expression

16 20 41 p U 0.014 128

38 103 18 p ¼ 0.760 44

44 111 7 p ¼ 0.354 27

48 125 8 p ¼ 0.986 21

32 64 23 p ¼ 0.064 83

31 84 17 p p ¼ 0.602 49 9 16

Claudin 3 expression

12 24 35 p ¼ 0.098 135

30 111 17 p ¼ 0.371 46

37 118 8 p ¼ 0.899 27

39 134 8 p ¼ 0.621 22

27 69 21 p ¼ 0.155 86

26 89 14 p ¼ 0.605 52 18 18

Claudin 4 expression

Focal expression, when up to 30% of the total area of the prostatic tissue was stained by the marker; overexpression, for staining of more than 30% of the prostatic tissue. Bold p values statistically significant.

*

Yes

Perineural invasion No

Yes

Biochemical recurrence No

Yes

Extraprostatic extension No

Yes

Seminal vesicle extension No 174

III/IV

97

26

8–9

Clinical stage I/II

66

116

Gleason score 4–6

7

No. patients

Correlation between claudin expression and clinicopathological characteristics of prostate adenocarcinomas

Characteristic

Table 4

15 21 44 p ¼ 0.060 125

40 100 18 p ¼ 1.000 45

46 108 8 p ¼ 0.407 27

49 123 8 p ¼ 0.838 22

30 65 29 p ¼ 0.485 78

33 81 16 p ¼ 0.228 50 11 15

Claudin 5 expression

36 0 156 p ¼ 0.075 14

132 9 58 p ¼ 0.685 5

146 9 32 p ¼ 0.544 3

163 10 27 p ¼ 0.383 3

93 3 97 p ¼ 0.071 10

110 5 58 p ¼ 0.110 8 25 1

Claudin 7 expression

146 COUTINHO-CAMILLO et al. Pathology (2011), 43(2), February

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CLAUDIN EXPRESSION IN PROSTATE ADENOCARCINOMAS

is consistent with Sheehan et al.,16 who found that decreased claudin 4 expression in prostate carcinoma is associated with advanced stages of tumours. Other groups also investigated claudin 4 expression in colorectal carcinoma and noted decreased expression in 57% of the tumours, especially in invasive front and metastatic colorectal carcinoma.20 Although claudin 4 was underexpressed in our neoplastic prostate samples, we could not link this phenomenon with clinicopathological parameters. Additionally, our results did not support the relationship between loss of claudin 4 expression and histological differentiation, because decreased claudin 4 expression in neoplastic samples did not correlate with Gleason scores in our series. In contrast to our findings and those of other studies, Landers et al.19 reported claudin 4 up-regulation in primary and metastatic prostate tumour specimens at the RNA and protein levels. They also correlated lower Gleason scores with higher claudin 4 expression. We observed significantly elevated expression of claudins 3 and 5 in neoplastic prostate tissue and moderate overexpression of claudin 2. Dysregulation of adhesion molecules can also manifest as increases in their expression in several types of neoplasms, and recent studies have shown that claudin overexpression activates TCF-LEF/b-catenin, an important cell cycle control pathway. This complex acts as a transcription factor that induces the expression of oncogenes that affect cell proliferation, survival, and invasion (myc, cyclin D1, MMP-7).21 Claudins 1 and 7 had similar expression profiles in neoplastic and BPH tissues in our TMA samples. The expression of these claudins varies, and studies of prostate tissue have described disparate expression profiles of these molecules in other series. Va¨re et al.18 reported strong immunoreactivity of claudin 1 in prostate adenocarcinoma. In contrast, Krajewska et al.17 found that claudin 1 was absent in 98% (47/48) of prostate carcinoma samples, whereas in BPH pronounced claudin 1 expression was observed. In other tumours, conflicting data on claudin 1 and 7 expression also exist. Downregulation of claudin 1 was reported in breast tumours22 and colon cancer9 and was shown to have a prognostic value. However, Lee et al.,23 noted a direct relationship between increased expression of claudins 1 and 7 and worsening histopathological grade in cervical carcinoma. Regarding claudin 7, in contrast to our findings, Sheehan et al.16 observed decreased expression in prostate adenocarcinoma compared with benign prostatic epithelium. Despite these differences in claudin 7 expression between series of prostate carcinoma, other groups have hypothesised that claudin 7 regulates the physiology of the prostate, because its expression is responsive to androgen stimulation in vitro.24 This finding suggests that claudin 7 is up-regulated by androgens and that its increase subsequently induces PSA gene expression, among others. However, our results failed to confirm this in vitro evidence, because expression of claudin 7 in prostate carcinoma samples did not differ statistically from BPH tissue, nor did it correlate with PSA levels. Claudin 11 was not expressed in neoplastic or BPH prostate tissue in our TMA samples. Some studies have shown that claudin 11 is expressed in brain and testis.25,26 Yet, no other study has analysed its expression in prostate samples.

147

Our analysis of clinicopathological parameters showed that only increased claudin 3 expression in prostate carcinoma specimens was marginally associated with advanced stages of disease. Additionally, higher expression of claudins 3 and 5 was associated with the presence of tumour perineural invasion of the specimens. The presence of extensive perineural invasion in prostate biopsies can predict extraprostatic tumour invasion and lymph node metastasis. In some studies, this standard, with Gleason histological grade, is an independent prognostic factor.27 Some studies have investigated the prognostic significance or perineural invasion in radical prostatectomy specimens, however the results remained controversial, due to the difficulty of a quantitative measurement of nerve invasion in pathological samples and the superiority of other well-established prognostic factors.28,29 Despite the notable relationship between claudins 3 and 5 and certain clinicopathological parameters of tumour evaluation, the claudins that we investigated showed no statistical correlation with Gleason score or biochemical recurrence of the disease. Krajewska et al.,17 using the same methods, did not report any association between claudins and the clinicopathological features of prostate adenocarcinoma, although a loss of claudin 1 was detected in 98% of their cases. However, in other series, reduced overall expression of claudins, especially of claudins 1 and 5, was associated with tumours that had a high Gleason score, suggesting that claudins modulate the histological features of these neoplasms and influence tumour behaviour.18 In summary, although claudin expression patterns were reported here and by other groups, they differ substantially depending on the series and the experimental approach. These differences are likely to be due to the great variability in genetic mechanisms that are involved in cancer biology. The molecular role of claudins in this process is largely unknown and merits further investigation to understand the biological significance of altered claudin expression in prostate cancer. Acknowledgements: FAPESP 98/14335-2. Address for correspondence: Mrs C. M. Coutinho-Camillo, Departamento de Anatomia Patolo´gica, Hospital AC Camargo, Rua Professor Antonio Prudente, 109 18 Andar, Sa˜o Paulo 01509-010, Brazil. E-mail: [email protected]

References 1. Jemal A, Siegl R, Ward E, et al. Cancer Statistics, 2007. CA Cancer J Clin 2007; 57: 43–66. 2. INCA (Instituto Nacional do Caˆncer), Ministe´rio da Sau´de. Estimativa de Incideˆncia de Caˆncer para 2008. Rio de Janeiro: INCA, 2008. www.inca. org.br. 3. Morin PJ. Claudin proteins in human cancer: promising new targets for diagnosis and therapy. Cancer Res 2005; 65: 9603–6. 4. Anderson JM. Molecular structure of tight junctions and their role in epithelial transport. News Physiol Sci 2001; 16: 126–30. 5. Gonzalez-Mariscal L, Betanzos A, Nava P, et al. Tight junction proteins. Prog Biophys Mol Biol 2003; 81: 1–44. 6. Furuse M, Fuita K, Hiiragi T, et al. Claudin-1 and -2: novel integral membrane proteins localizing at tight junctions with no sequence similarity to occluding. J Cell Biol 1998; 141: 1539–50. 7. Kominsky SL. Claudins: emerging targets for cancer therapy. Expert Rev Mol Med 2006; 8 (18): 1–11. 8. de Oliveira SS, de Oliveira IM, De Souza W, et al. Claudins upregulation in human colorectal cancer. FEBS Lett 2005; 579: 6179–85. 9. Resnick MB, Konkin T, Routhier J, et al. Claudin-1 is a strong prognostic indicator in stage II colonic cancer: a tissue microarray study. Mod Pathol 2005; 18: 511–8.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

148

COUTINHO-CAMILLO et al.

10. Michl P, Buchholz M, Rolke M, et al. Claudin-4: a new target for pancreatic cancer treatment using Clostridium perfringens enterotoxin. Gastroenterology 2001; 121: 678–84. 11. Nichols LS, Ashfaq R, Iacobuzio-Donahue CA. Claudin 4 protein expression in primary and metastatic pancreatic cancer: support for use as a therapeutic target. Am J Clin Pathol 2004; 121: 226–30. 12. Hoevel T, Macek R, Mundigl O, et al. Expression and targeting of the tight junction protein CLDN1 in CLDN1-negative human breast tumor cells. J Cell Physiol 2002; 191: 60–8. 13. Kominsky SL, Argani P, Korz D, et al. Loss of the tight junction protein claudin-7 correlates with histological grade in both ductal carcinoma in situ and invasive ductal carcinoma of the breast. Oncogene 2003; 22: 2021–33. 14. Agarwal R, D’Souza T, Morin PJ. Claudin-3 and claudin-4 expression in ovarian epithelial cells enhances invasion and is associated with increased matrix metalloproteinase-2 activity. Cancer Res 2005; 65: 7378–85. 15. Rangel LB, Agarwal R, D’Souza T, et al. Tight junction proteins claudin-3 and claudin-4 are frequently overexpressed in ovarian cancer but not in ovarian cystadenomas. Clin Cancer Res 2003; 9: 2567–75. 16. Sheehan GM, Kallakury BVS, Sheenan CE, et al. Loss of claudins-1 and -7 and expression of claudins-3 and -4 correlate with prognostic variables in prostatic adenocarcinomas. Hum Pathol 2007; 38: 564–9. 17. Krajewska M, Olson AH, Mercola D, et al. Claudin-1 immunohistochemistry for distinguishing malignant from benign epithelial lesions of prostate. Prostate 2007; 67: 907–10. 18. Va¨re P, Loikkanen I, Hirvikoski P, et al. Low claudin expression is associated with high Gleason grade in prostate adenocarcinoma. Oncol Rep 2008; 19: 25–31.

Pathology (2011), 43(2), February

19. Landers KA, Samaratunga H, Teng L, et al. Identification of claudin-4 as a marker highly overexpressed in both primary and metastatic prostate cancer. Br J Cancer 2008; 99: 491–501. 20. Ueda J, Semba S, Chiba H, et al. Heterogeneous expression of claudin-4 in human colorectal cancer: decreased claudin-4 expression at the invasive front correlates cancer invasion and metastasis. Pathobiology 2007; 74: 32–41. 21. Dhawan P, Singh AB, Deane NG, et al. Claudin-1 regulates cellular transformation and metastatic behavior in colon cancer. J Clin Invest 2005; 115: 1765–76. 22. Toke´s AM, Kulka J, Paku S, et al. Claudin-1, -3 and -4 proteins and mRNA expression in benign and malignant breast lesions: a research study. Breast Cancer Res 2005; 7: R296–305. 23. Lee JW, Lee SJ, Seo J, et al. Increased expressions of claudin-1 and claudin-7 during the progression of cervical neoplasia. Gynecol Oncol 2005; 97: 53–9. 24. Zheng JY, Yu D, Foroohar M, et al. Regulation of the expression of the prostate-specific antigen by claudin-7. J Memb Biol 2003; 194: 187–97. 25. Lui WY, Lee WM, Cheng CY. Transforming growth factor beta3 regulates the dynamics of Sertoli cell tight junctions via the p38 mitogen-activated protein kinase pathway. Biol Reprod 2003; 68: 1597–612. 26. McCullumsmith RE, Gupta D, Beneyto M, et al. Expression of transcripts for myelination-related genes in the anterior cingulated cortex in schizophrenia. Schizophr Res 2007; 90: 15–27. 27. De Torres Ramirez I. Prognosis and predictive factors of prostate cancer in the prostatic biopsy. Actas Urol Esp 2007; 31: 1025–44. 28. Marchesi F, Piemonti L, Mantovani A, Allavena P. Molecular mechanisms of perineural invasion, a forgotten pathway of dissemination and metastasis. Cytokine Growth Factor Rev 2010; 21: 77–82. 29. Masieri L, Lanciotti M, Nesi G, et al. Prognostic role of perineural invasion in 239 consecutive patients with pathologically organ-confined prostate cancer. Urol Int 2010; May 28: [Epub ahead of print].

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.